Agent Browser

Name: agent-browser
Rating: 88
Author: mikeparcewski

Use the agent-browser CLI for headless browser automation.

Prerequisites

bash

# Check if installed
which agent-browser

# Install via npm
npm install -g agent-browser

# Or use npx (no install)
npx agent-browser --help

Core Commands

bash

# Open a URL
agent-browser open https://example.com

# Take screenshot
agent-browser screenshot --output screenshot.png

# Get page snapshot (accessibility tree + content)
agent-browser snapshot

# Click element
agent-browser click "button.submit"

# Fill form field
agent-browser fill "input[name=email]" "user@example.com"

# Extract text content
agent-browser text "main"

# Get page HTML
agent-browser html

Common Workflows

Screenshot a Page

bash

agent-browser open https://example.com
agent-browser screenshot --output page.png

Fill and Submit Form

bash

agent-browser open https://example.com/login
agent-browser fill "#email" "user@example.com"
agent-browser fill "#password" "secret"
agent-browser click "button[type=submit]"
agent-browser wait 2000
agent-browser screenshot --output result.png

Extract Page Content

bash

agent-browser open https://example.com
agent-browser snapshot  # Returns accessibility tree + text

Accessibility Audit

bash

agent-browser open https://example.com
agent-browser a11y  # Returns accessibility violations

Live Preview

Stream the browser viewport via WebSocket:

bash

agent-browser open https://example.com --stream
# Opens ws://localhost:9222 for live viewport

Quick Reference

Command	Purpose
`open URL`	Navigate to URL
`screenshot`	Capture viewport
`snapshot`	Get accessibility tree + content
`click SELECTOR`	Click element
`fill SELECTOR VALUE`	Fill input field
`text SELECTOR`	Extract text content
`html`	Get page HTML
`a11y`	Accessibility audit
`wait MS`	Wait for milliseconds

For full reference: https://github.com/vercel-labs/agent-browser